Comparative Transcriptome Analysis of Storage Root Reveals Regulatory Mechanisms of Lignin Biosynthesis During Root Development in Two Sweetpotato Cultivars

doi:10.21203/rs.3.rs-850254/v1

Background

Sweetpotato(Ipomoea batatas (L.) Lam.) is one of the most important crops with high storage roots yield. Lignin affects the storage root formation. However, the molecular mechanisms of lignin biosynthesis in storage roots development have been lacking.

Results

To reveal the molecular mechanism of lignin biosynthesis and identify new homologous genes in lignin biosynthesis during storage root development, the storage root (SR) at three different stages (D1, D2 and D3) in the two cultivars (Jishu25 and Jishu29) was investigated with full-length and second-generation transcriptome. A total of 52,137 transcripts and 21,148 unigenes were obtained after corrected with Hiseq2500 sequencing. Through the comparative analysis, 9577 unigenes were found to be differently expressed in the different stage in two cultivars. Among of them, 91 unigenes enriched in the phenylpropanoid biosynthesis, and 201 unigenes in hormone signal transduction pathway with KEGG analysis. Weighted gene co-expression network analysis of differentially expressed unigenes showed that lignin biosynthesis genes might be co-expressed with transcription factors such as AP2/ERF and MYB at the transcription level, and regulated by phytohormones auxin and GA₃.

Conclusions

Taken together, our findings will throw light on molecular regulatory mechanism of lignin biosynthesis involved in storage root development.

Plant Molecular Biology and Genetics

Plant Physiology and Morphology

Sweetpotato

Storage root

Lignin biosynthesis

Hormone

Transcription factors

Sweetpotato (Ipomoea batatas (L.) is one of the seventh important food crops in the world, and produces approximately 112.9 million tons storage root from the area of 9.1 million ha (FAO stat, 2018). China is the largest sweetpotato producing country with 70.3 million tons, which accounts for 62.3% of the total production in sweetpotato worldwide. Storage root (SR) is the main edible tissue, its formation and development is the most important agronomic trait in sweetpotato producing, which the complex biological process is involving in morphogenesis as well as carbohydrate, storage proteins, and secondary metabolites accumulation from adventitious roots [1]. SR yield varied in sweetpotato cultivars, and is prone to environmental change [2, 3, 4]. It was reported that SR yield is not only dependent upon the rate and duration of SR expansion, but also on the beginning of SR formation, leaves longevity and the growth stage [5, 6, 7].

Lignin biosynthesis is one of phenylpropanoid compounds and plays an important role in the secondary cell walls formation [8], so be considered the important roles in storage root formation of sweetpotato [9, 10]. Lignin polymer generally comprises p-hydroxyphenyl (H), guaiacyl (G) and syringyl (S) units in plants, which be synthesized by phenylalanine by a series of enzymes, such as cinnamate 4-hydroxylase (C4H), 4-coumarate CoA ligase(4CL), 4-hydroxycinnamoyl-CoA shikimate/quinate4-hydroxycinnamoyl transferase (HCT), caffeoyl-CoA O-methyltransferase (CCoAOMT), cinnamoyl-CoA reductase (CCR), ferulate 5-hydroxylase (F5H), caffeic acid 3-O-methyltransferase (COMT), cinnamylalcohol dehydrogenase (CAD), and peroxidases (PER). The lignin biosynthetic genes are transcriptionally regulated by the transcriptional regulators. Transcription factors (TF) in the regulation of lignin biosynthesis genes also include NAC [11], MYB [12], WRKY [13]. Recent studies revealed that miRNA is involved in the regulation of lignin biosynthesis [14, 15]. Jin [16] found that calcium-dependent protein kinase (CDPK) mediated methionine adenosyl transferase via 26S proteasome pathway and affects ethylene biosynthesis and lignin deposition in Arabidopsis. Gibberellin (GA₃) and cytokinin are reported to be involved in regulating lignin biosynthesis [17, 18].

Studies have shown that lignin biosynthesis is connected with root formation and development in sweetpotato. Firon [10] compared the expression profiles of initiating storage roots and fibrous roots, and identified the lignin biosynthesis down-regulated at early stage of storage root formation. Wang [9] found that ectopic expression of maize Lc regulatory gene in sweetpotato induced the expression of lignin biosynthesis genes and affect storage root development. Kim [19] discovered many differently expressed genes related to phenylpropanoid biosynthesis in adventitious root formation through RNA-seq analysis. However, the molecular mechanism of lignin in SR development keeps obscures.

In this study, we selected SR at three developmental stages of two sweetpotato varieties, compared the full-length transcriptome data and investigated the gene expression profiles by using full-length and second-generation transcriptome. The primary objective is to reveal the mechanism of lignin biosynthesis, and identify the important genes required for lignin biosynthesis in the SR development of sweetpotato.

1. Characteristic of storage root development at different stages in two cultivars

SRs of Jishu25 and Jishu29 at three developmental stages were harvested and investigated (Fig.1A). From D1 to D3, R/S and the SR numbers in cv.Jishu29 were all significantly higher than those of Jishu25 (Fig.1B and Fig.1C). Compared with cv. Jishu25, the root yield in cv. Jishu29 was increased (Fig.1D). Meanwhile, the root expansion rate in Jishu29 was faster than that of Jishu25 during the early stage.

2. Full-length mRNA sequencing

To compare the molecular mechanisms of tuberous root development of both cultivars, three cDNA libraries from each cultivar prepared from the middle section of storage roots at three stages were subjected to the second-generation transcriptome sequencing in three biological replicates (Tabel 1). Meanwhile, the RNA of each sample was mixed for SMRT sequencing. We obtained 741,084 circular consensus sequencing (CCS) reads ranging from 51bp to 14,896bp with an average length of 1,322bp, which included 77.1% of full-length non-chimeric (FLNC) and 20.7% of non-full-length (NFL) reads (Additional file 1: Table S1). After corrected with Hiseq2500 sequencing, 52,137 transcripts and 21,148 unigenes were obtained with an average length of 1,253bp and 1,457bp, which indicate that the data are of high quality (Additional file 2: Table S2). The sequence length after redundancy varied from 53bp to 6,868bp with the mean length of 1457bp.

3. Annotation and classification of sweetpotato unigenes

To identify the predictive functions of unigenes in sweetpotato storage roots, all of the assembled unigenes were matched against the seven databases. Based on the sequence similarity, a total of 96.51% of the unigenes were annotated by alignment in at least one database. Of them, 19,137 (90.49%) unigenes were aligned against the NR databases, 16,083 (76.05%) against SwissProt, 18,849 (89.13%) against KEGG, 11,585 (54.78%) against KOG, 12,949(61.23%) against GO, 20,114 (95.11%) against NT, and 12,949 (61.23%) against Pfam (Table 2).

In GO classification, 12494 unigenes were successfully assigned to 51 functional groups, of which 25 groups belonged to the biological process, 10 to molecular function and 16 to cellular component. In the category of biological process, the most abundant groups contained metabolic process (6413 unigenes, 51.33%) and cellular process (6198 unigenes, 49.61%). For molecular function, the highest categories were binding (7133 unigenes, 57.09%) followed by catalytic activity (5943 unigenes, 47.57%). Furthermore, the majority of cellular components were cell (2639 unigenes, 21.12%) and cell part (2639 unigenes, 21.12%) (Additional file 3: Fig.S1).

For KEGG annotation, 18,849 unigenes were clustered into 44 subcategories. As shown in Fig.S2 (Additional file 4), the groups “the signal transduction pathways” (932 unigenes, 4.94%) and “the translation pathways” (840 unigenes, 4.46%) formed the two largest clusters, followed with the energy metabolism (796 unigenes, 4.22%) and the carbonydrate metabolism (785 unigenes, 4.16%).

After KOG annotation, 11,585 unigenes were divided into 24 functional clusters as shown in Fig.S3 (Additional file 5). The top three categories included general function prediction only (1885 unigenes, 16.27%), posttranslational modification, protein turnover, chaperones (1614 unigenes, 13.93%) and signal transduction mechanisms (1185 unigenes, 10.23%).

4. Analysis of differentially expressed unigenes (DEG)

In this study, we analyzed the global gene expression profiles of sweetpotato storage root in different developmental stages. According to the false discovery rate and fold change, 14,598 DEGs were screened through cluster analysis in the storage roots during the three different developmental stage of two sweetpotato cultivars.

To examine the gene expression difference during the root development stages in the two genotypes, the DEGs were identified by the comparisons of the nine DEG libraries i.e. J25-D2 vs J25-D1, J25-D3 vs J25-D2 (Additional file 6: Fig.S4). The largest number of DEGs was occurred between J29-D3 vs J25-D3 with 3063 up-regulated and 4304 down-regulated unigenes. Furthermore, 6675 and 6571 unigenes were significantly differentially expressed between J25-D3 vs J25-D2, and J29-D1 vs J25-D1, respectively. It is interesting that the number of DEGs marked in the two genotypes increased from D1-D2 comparison to D2-D3 (Fig.2). Meanwhile, the number of up- or down-regulated DEGs in Jishu25 was much bigger than those in cv.Jishu29, which revealed that the transcriptome of Jishu25 changed drastically compared to Jishu29 (Fig.2).

5. Pathway analysis of differentially expressed genes

To find out the correlation between phenotypic differences and gene expression, KEGG pathway enrichment of DEGs were performed in J29-D3 vs J25-D3, J29-D1 vs J25-D1 and J25-D3 vs J25-D2. The DEGs in J29-D3 vs J25-D3 were identified to be involved in plant-pathogen interaction (98 unigenes) and phenylpropanoid biosynthesis (60 unigenes) pathway (Fig.3A). Of them, the down-regulated unigenes were enriched in plant-pathogen interaction (86 unigenes), phenylpropanoid biosynthesis (54 unigenes), amino sugar and nucleotide sugar metabolism (46 unigenes). In J29-D1 vs J25-D1, the DEGs were significantly enriched in plant hormone signal transduction pathways (113 unigenes) (Fig.3B). The DEGs in J25-D3 vs J25-D2 were significantly expressed in starch and sucrose metabolism (73 unigenes), plant hormone signal transduction (108 unigenes), and porphyry and chlorophyll metabolism pathway (39 unigenes) (Fig.3C).

6. Differentially expressed unigenes in lignin biosynthesis

In cv. Jishu29, there are 64 and 40 DEGs in phenylpropanoid biosynthesis pathways in the transcriptome data of J29-D3 vs J29-D1 and J29-D3 vs J29-D2. We identified 60 unigenes enriched in phenylpropanoid biosynthesis pathways from the comparison data of J29-D3 vs J25-D3. Analysis of differentially expressed unigenes between them showed that 4 unigenes were differentially expressed in the coumarine biosynthesis, and 56 unigenes in lignin biosynthesis, of which 6 were up-regulated and 50 were down-regulated. These genes include phenylalanine ammonia-lyase (PAL), trans-cinnamate 4-monooxygenase (CYP73A), cinnamate 4-monooxygenase (C4H), 4-coumarate--CoA ligase(4CL), cinnamoyl-CoAreductase (CCR), shikimate O-hydroxy cinnamoyl transferase (HCT), cinnamyl-alcohol dehydrogenase (CAD), caffeic acid 3-O-methyltransferase (COMT), caffeoyl-CoA O-methyltransferase (CCoAMT), and peroxidase (PER) ( Additional file 7:Table S3).

7. Differentially expressed unigenes in hormone signal transduction pathway

Through comparisons between the three stages in cv. J25, 84 down-regulated and 24 up-regulated unigenes in hormone signal transduction pathway were found to be differentially expressed in D3 vs D2, 52 up-regulated and 47 down-regulated unigenes were differentially enriched in D2 vs D1 (Table 3). After comparisons between the three stages in cv. J29, 37 down-regulated unigenes in hormone signal transduction pathway were found to be differentially expressed in D3 vs D2. The unigenes after KEGG analysis were significantly enriched in signaltransduction pathways such as auxin, cytokinine, abiscisic acid, ethylene, jasmonic acid, salicylic acid and brassinosteroid. Meawhile, we found that a total of 113 unigenes were differentially expressed in J29-D1vsJ25-D1 (Table 2).

8. Differentially expressed transcription factors in root development

A total of 622 transcription factors (TF) were differently expressed in J29-D3 vs J25-D3. Among them, the largest types were AP2/ERF-ERF (74), bHLH (40), NAC (37), and WRKY (36)（Additional file 8: Fig S5A）,which were reported to related with root formation and development [20, 21, 22]. NAC and WRKY TF are the key regulators of the lignifications of vessel cell differentiation [23, 24]. In addition, 29 GRAS and 16 AUX/IAA transcription factors, which play a crucial role in gibberellins and auxin signal transduction [25, 26], were found. There are a total of 570 TFs differently expressed in J29-D3 vs J25-D2. Of which, 292 significantly up-regulated and 278 were significantly down-regulated (Additional file 8: Fig S5B).

To identify the DEGs correlating with tuberous root formation and development, WGCNA was implemented to construct gene network from 27 sweetpotato root samples in two cultivars by R package [27]. In the analysis, 17 stable co-expressed modules were obtained through WGCNA (Fig.4A). In the brown module, GO enrichment analysis showed that the unigenes were mainly involved in transferase activity and transcription factor activity in molecular function category, S-adenosylmethionine biosynthetic and metabolic process in biological process category, and apoplast in cellular component (Fig.4B). KEGG enrichment analysis showed that the unigenes were significantly enriched in phenylpropanoid biosynthesis, and plant-pathogen interaction (Fig.4C).

10. Validation of RNA-seq results

To validate our transcriptome data, qRT-PCR analyses were performed to determine the expression of 12 random DEGs in three root developmental stages of two cultivars (primers in additional file 9).The qRT-PCR results showed that the expression patterns of the 12 DEGs were in good agreement with their RNA-seq results in the root development stages in every cultivar (Fig 5), and the positive correlation coefficient (R²) was 0.9195. Therefore, the transcriptome data was highly reliable.

11. Change in lignin content in root development

Lignin affects storage root development in sweetpotato. To understate the dynamic changes in lignin accumulation in sweetpotato, the lignin content in SR were detected during the root development stages in two cultivars. The results showed that lignin content in SR decreased gradually from D1 to D3 stages in the two cultivars, and the lignin content in cv.Jishu29 was significantly higher than those of Jishu25 at D1and D2 stage (Fig.6A).

To clarify the molecular mechanism underlying lignin in root development, the expression patterns of six lignin biosynthesis-related genes were determined at root developmental stages in the two cultivars. In cv.Jishu25, the genes Ib4CL, IbCAD, IbCCoAMT, IbCCR and IbCOMT was expressed highly at D1 stage, and then decreased at D2 and D3 stage, which were consistent with lignin accumulation and the transcriptome obtained from Jishu25 (Additional file 10: Table S5). IbPER displays the highest mRNA at D2 stage. In cv. Jishu29, their expression patterns of Ib4CL, IbCAD, IbCCoAMT and IbCOMT were different to those in Jishu25. The transcript levels of Ib4CL, IbCAD, IbCCoAMT, and IbCOMT reach the peak at 46 DAT. Transcription of IbPER showed a pattern similar to that of IbCCR, which reduced in the three developmental stages. The expressional profiles of these two genes were correlated with lignin accumulation in Jishu29 (Fig.6B).

Storage root（SR）formation and development are important for sweetpotato production. SR formation includes the initiation of vascular cambium and several irregular circular cambiums around the primary and secondary xylem. Root lignifications directly inhibit SR formation [28]. Firon [10] and Wang [9] discovered that many structural genes related to lignin biosynthesis through transcriptome analysis down-regulated in SR formation. SR development is due to the increase in the number of secondary xylem and phloem cells by anomalous cambia cell division and proliferation. The competition between lignifications and cambia activity, and SR growth rate are the important factors to determine SR yield [28, 29]. However, lignin in SR development has not yet been reported in sweetpotato. In this study, we compared six RNA-sequence datasets of three root development stages in Jishu29 and Jishu25, which are different in SR number and SR growth rate, and found that lignin content decreased and its biosynthesis genes displayed the different expression patterns in root development through transcriptome analysis. Siebers [30] found that there is a correlation between lignin biosynthesis regulators and fibrous root development in cassava. To our knowledge, this work is the first to systematically report the role of lignin in SR development of sweetpotato.

In this study, the expression transcripts of 6 genes were investigated to study the roles in lignin biosynthesis during SR development of sweetpotato. 4 genes (IbCCoAMT, IbCCR, IbPER and IbCOMT) were associated with lignin biosynthesis. Among them, the expression levels of IbPER and IbCCR were positively correlated with lignin accumulation during storage root development indicated that they might play an important role lignin biosynthesis in sweetpotato. The cinnamoyl-CoA reductase (CCR) catalyzes the most important metabolic reaction in the monolignol biosynthesis by transferring phenylpropionic metabolites into the lignin synthesis pathway. Down-regulation of CCR gene in warm seasonforage grass Paspalum dilatatum. And poplar significantly reduced lignin content [31, 32]. PER is the key enzyme involved in the monomers polymerization. Over-expression PER gene (swPA4) in tobacco increase the lignin [33] and down-regulation of TP60 decreased lignin level [34]. Wang [35] found that DcPER1 transcript level was consistent with the change in lignin content in carrot. A similar result was observed during SR development in sweetpotato.

Increasing evidence showed that phytohormone is associated with changes in lignifications [36, 37, 38].The lignin repressor PtMYB21a is upregulated by both IAA and GA₃ but down regulated by cytokinin in hybridaspen [36]. Herrero [39] reported that phytohormones such as auxins, cytokinins, BRs and GA₃might control AtPrx52 involved in lignin biosynthesis in Arabidopsis thaliana. Auxin was considered to be involved in cell expansion and cell wall biosynthesis [37, 40]. Ravi and Indira [1] reported that low concentration of IAA and high activity of IAA oxidase were related to the lignification of storage root of sweetpotato. Tamaoki [38] revealed that the expression of lignin biosynthesis related genes increased with the auxin inflow increase of the shoot apex in Arabidopsis. Here, the expression levels of the unigenes encoding auxin-responsive protein (transcript26216) and auxin response factor (transcript1390) were found to be down-regulated in the low-lignin culitivar. GA₃ application enhanced lignin accumulation and decreased starch content in sweetpotato roots, which inhibited SR formation [41]. The similar results were found in carrot [35]. Transcriptome analysis showed that GID1 (transcript4234) as the gibberellin receptor was significantly down regulated expression in the low lignin cultivar. Based on these results, Auxin interacts with other hormones GA₃ to function lignin accumulation during storage root development in sweetpotato.

Lignin biosynthesis in plants is controlled by a regulatory network. Transcription factors include AP2/ERF (transcript11979) and MYB-CC (transcript19170) and KNOTTED-like homeobox (transcript29522), as a transcriptional activator, promoter the lignin synthesis by regulating the expression of C4H, CCoAOMT1. Ma [42] reported that Ii049, the AP2/ERF TF, regulated lignin biosynthesis in Isatisindigotica. The over-expression of GbERF1 in cotton resulted in the increase of lignin content [43]. Our transcriptome and qRT-PCR results suggested that the transcript level AP2/ERF (trancript30056) decreased in SR development, which was consistent with the change trend of the lignin content, indicated that ERF might act as the regulator in lignin biosynthesis during the development of storage roots. The regulator containing MYB3 (transcript25828), as a transcriptional inhibitor, regulate negatively the lignin accumulation. IbMYB may negatively regulate lignin biosynthesis via binding miR828 in sweetpotato on wounding [44]. Taken together, these results indicated that AP2/ERF and MYB transcriptional factors might be related to lignin biosynthesis.

Plant material

Sweetpotato cultivars “Jishu25 (J25)” and “Jishu29 (J29)” were planted in the experimental station of the Crops Research Institute, Shandong Academy of Agricultural Sciences, Jinan, China. Storage roots at the three stages were collected from sweetpotato plants at 32(D1), 46(D2), 67(D3) days after transplantation as detailed in Villordon [45] methods, respectively. The samples were immediately frozen in liquid nitrogen and stored at -80 °C for hormone and RNA isolation. At the root developmental stages (At the same developmental stage of the roots), the numbers and yield of storage root was counted and weighed. The root/shoot (R/S) ratio was calculated according to the root weight divided by the shoot biomass of per plant.

RNA extraction, Full-length cDNA library construction and sequencing

Total RNAs was extracted using RNAprep Pure Plant plus Kit (Tiangen Biotech (Beijing) CO. LTD.) and purified with the RNAeasy Plant Mini Kit (Qiagen, Valencia, CA). RNA quality was verified using a 2100 Bioanalyzer RNA Nanochip (Agilent, Santa Clara, CA), and been quantified using NanoDrop ND-1000 Spectrophotometer (Nano-Drop, Wilmington, DE).

For full-length cDNA library, cDNA was synthesized using SMARTer PCR cDNA Synthesis Kit, and optimized for PCR amplification. The fragments for large-scale PCR were performed using magnetic beads to obtain sufficient total cDNA. The complete SMRT bell library was constructed with using a SMARTer PCR cDNA Synthesis Kit and assembly was performed on the PacBio Sequel platform, the second-generation sequencing and assembly was implemented on the Hiseq 2500 sequencing platform (Illumina) with PE150 by Novogene Co., LTD (Beijing, China).

Functional annotation, identification and analysis of DEGs

The full cDNA sequence was processed with SMRTlink7.0 software and corrected with the second-generation transcription data to obtain consistent unigenes, then removed the any redundancy unigenes by CD-HIT software [46]. Unigenes were functionally annotated using the BLASTX alignment (E-value ≤ 10^-5) against seven databases, including GO (Gene Ontology), KEGG (Kyoto Encyclopedia of Genes and Genomes), KOG (Eukaryotic Ortholog Groups), Nr (NCBI nucleotide sequences), Nt (NCBI nucleotide sequences), Pfam (Protein Family Database)and Swiss-prot (a manually annotated and reviewed protein sequence database) databases. The best alignment results were selected for the annotation of the unigenes.

Gene expression levels were analyzed by RSEM software [47]. Differential expression genes (DEGs) among storage roots of the different developmental stage was identified by DESeq package (1.10.1) with |log₂(FoldChange)| > 0 and qvalue< 0.05 [48]. GO enrichment analysis of DEGs was performed using Goseq softberry with corrected and pvalue<0.05. KEGG enrichment was analyzed by KOBAS3.0 softberry with corrected pvalue<0.05. To further analyze the regulatory mechanism of root formation and development, weighted gene co-expression networks analysis (WGCNA) was performed with the WGCNA package in the R software [49], and network visualization for each module was analyzed using the Cytoscape software 3.6.1[50].

Real-time quantitative PCR validation

To confirm the expression of unigenes, 12 unigenes were selected for qRT-PCR. The analysis was performed using samples with tuberous roots at the three stages from two cultivars. The qRT-PCR and data analyses were performed as described by Qin [51]. 12 unigenes, including 4 plant hormone biosynthesis-related genes, 5 lignin biosynthesis-related genes, and 3 transcription factors from the RNA-Seq were validated, and the primers used for the validation were listed in Table S4. Sweetpotato Ibactin was used as the reference gene for normalizing quantities of gene expression.

Measurements of dry matter and hormone contents

To measure the root dry matter content, a freshly mixed 20 g sample of root slices was oven-dried to a constant weight at 80°C for 48 h. Meanwhile, 2 g fresh roots were used for IAA, ABA, GA₃, ZR and JA extractions and purification according to the Dobrev and Vankova’s method [52]. The hormonal quantification was performed using HPLC (High Performance Liquid Chromatography) as described by Djilianov [53], with no less than three independent biological replicates.

Determination of lignin content

Analysis of the lignin content was performed with slight modify according to the methods described by Van-Acker [54]. In short, 10 mg dried roots (W) were digested in 2.5 ml HoAc solution containing 25% (v/v) acetylbromide and 0.1 ml 70% perchloric acid. The sealed vessel was mixed fully and incubated for 40 min at 80°C while shaking. After incubation and cooling, the slurry was centrifuged at 23,477 g for 15 min. The supernatant was added to 2.5 ml of 2 M NaOH and 1 ml acetic acid. After 20 min, the absorbance (A) was measured at 280 nm. The lignin content was calculated using the following formula: lignin (mg/g) = 0.147× (ΔA-0.0068) ÷W×50.

Statistical analysis

Statistical analysis was performed using SPSS version13.0 with ANOVA and Duncan’s test for dry matter, lignin, hormone and qPCR results. Data are means with three biological replicates, with each error bar representing standard error. The statistical significance difference was calculated with Duncan's new multiple ranges test and marked with asterisks at p< 0.05.

C4H: Cinnamate 4-hydroxylase; 4CL: 4-coumarate CoA ligase; CCoAOMT: caffeoyl-CoA O-methyltransferase; CCR: cinnamoyl-CoA reductase; COMT: caffeic acid 3-O-methyltransferase; CAD: cinnamylalcohol dehydrogenase; F5H: ferulate 5-hydroxylase, HCT: 4-hydroxycinnamoyl-CoA:shikimate/quinate4-hydroxycinnamoyl transferase; calcium-dependent protein kinase (CDPK)PER: Peroxidases; Gibberellin (GA₃) circular consensus sequencing (CCS); DEG: differentially expressed unigenes; FLNC: full-length non-chimeric; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; KOG: EuKaryotic Ortholog Groups; NFL: non-full-length reads; R/S: The root/shoot ratio; SR： Storage root; WGCNA: Weighted gene co-expression networks analysis.

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

All data generated in the current study are included in the paper and the supplementary files. The Transcriptome data has been submitted to the NCBI SRA database, and its accession numbers is PRJNA756699 (https://www.ncbi.nlm.nih.gov/sra/?term= PRJNA756699).

Competing interests

The authors declare they have no competing interests.

Funding

This research was supported by National Key R&D Program of China (2019YFD1001300, 2019YFD1001304), Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX19_2200), the China Agriculture Research System of MOF and MARA, Taishan industry leading talents project (LJNY202002) and Shandong agricultural improved seed project (2020LZGC004).

Authors' contributions

FYH, ZYL and QMW designed the research. FYH wrote the paper. FYH and ZQ performed some experiments and analyzed the data, TFD, AXL, SXD and YYZ analyzed the data. LMZ and DFM revised the paper. All authors have read and approved the manuscript.

Acknowledgements

Not applicable

[1] Ravi V, Indira P. Crop physiology of sweet potato. Hortic. Rev. 1999; 23:277–338.

[2] Solis JA, Villordon N., Baisakh D, LaBonte，Firon N. Effect of drought on storage root development and gene expression profile of sweetpotato under greenhouse and field conditions. Journal of the American Society for Horticultural Science. 2014; 139:317-324

[3] Laurie SM, Booyse M. Employing the GGE SREG model plus Elston index values for multiple trait selection in sweetpotato. Euphytica. 2015; 204(2):433-442.

[4] Gurmu F, Hussein S, Laing M. Genotype-by-environment interaction and stability of sweetpotato genotypes for root dry matter, β-carotene and fresh root yield. Open Agriculture, 2017; 2: 473-485.

[5] Kim YH, Park SC, Ji CY, Lee JJ, Jeong JC, Lee HS, Kwak SS. Diverse antioxidant enzyme levels in different sweetpotato root types during storage root formation. Plant Growth Regul. 2015; 75:155–164. DOI 10.1007/s10725-014-9940-x

[6] Ma J, Aloni R, Villordon A, Labonte D, Kfir Y, Zemach H, Schwartz A, Althan L, Firon N. Adventitious root primordia formation and development in stem nodes of ‘Georgia Jet’sweetpotato, Ipomoea batatas. AMERICAN JOURNAL OF BOTANY, 2015; 102( 7 ): 1040 – 1049.

[7] Bararyenya A, Tukamuhabwa P, Gibson P, Grüneberg W, Ssali R, Low J, Odong T, Ochwo-Ssemakula M, Talwana H, MwilaN Mwanga R. Continuous Storage Root Formation and Bulking in Sweetpotato. Gates Open Research. 2019; 3: 83 https://doi.org/10.12688/gatesopenres.12895.1).

[8] Vanholme R, Demedts B, Morreel K, Ralph J, Boerjan W. Lignin Biosynthesis and Structure. Plant Physiology. 2010; 153: 895–905.

[9] Wang H, Yang J, Zhang M, Fan W, Firon N, Pattanaik S, et al. Altered phenylpropanoid metabolism in the maize Lc-expressed sweet potato (Ipomoea batatas) affects storage root development. Sci. Rep. 2016; 6: 18645. doi: 18610.11038/ srep18645

[10] Firon N, LaBonte D, Villordon A, Kfir Y, Solis J, Lapis E. Transcriptional profiling of sweetpotato (Ipomoea batatas) roots indicates down-regulation of lignin biosynthesis and up-regulation of starch biosynthesis at an early stage of storage root formation. BMC Genomics. 2013; 14：460.

[11] Ohashi-Ito K.Oda Y,Fukuda H. Arabidopsis VASCULAR-RELATED NAC-DOMAIN6 directly regulates the genes that govern pro-grammed cell death and secondary wall formation during xylem differentiation. Plant Cell. 2010; 22:3461–3473.

[12] Zhou J, Lee C, Zhong R, Ye ZH. MYB58 and MYB63 are transcriptional activators of the lignin biosynthetic pathway during secondary cell wall formation in Arabidopsis. Plant Cell. 2009; 21: 248–266.

[13] Yang L, Zhao X, Yang F, Fan D, Jiang Y, Luo K. PtrWRKY19, a novel WRKY transcription factor, contributes to the regulation of pith secondary wall forma-tion in Populus trichocarpa. Sci. Rep. 2016; 6: 18643.

[14] Chen HC, Song J, Wang JP, Lin YC, Ducoste J, Shuford CM, Liu J, Li Q, Shi R, Nepomuceno A, Isik F, Muddiman DC, Williams C, Sederoff RR, Chiang VL. Systems biology of lignin biosynthesis in Populustrichocarpa: heteromeric 4-coumaric acid:coenzyme A ligase protein complex formation, regulation, and numerical modeling. Plant Cell. 2014; 26: 876–893.

[15] Lu S, Li Q, Wei H, Chang MJ, Tunlaya-Anukit S, Kim H, Liu J, Song J, Sun YH, Yuan L, Yeh TF, Peszlen I, Ralph J, Sederoff RR, Chiang VL. Ptr-miR397a is a negative regulator of laccase genes affecting lignin content in Populustrichocarpa. Proc. Natl. Acad. Sci. U.S.A. 2013; 110: 10848–10853.

[16] Jin Y, Ye N, Zhu F, Li H, Wang J, Jiang L, Zhang J. Calcium-dependent protein kinase CPK28 targets the methionine adenosyltransferases for degradation by the 26S proteasome and affects ethylene biosynthesis and lignin deposition in Arabidopsis. The Plant Journal. 2017; 90:304–318. doi: 10.1111/tpj.13493

[17] Huang D, Wang S, Zhang B, Shang-Guan K, Zhou Y. A gibberellin-mediated DELLA-NAC signaling cascade regulates cellulose synthesis in rice. Plant Cell. 2015; 27:1681–1696.

[18] Zhang K, Novak O, Wei Z, Gou M, Zhang X, Yu Y , Yang H, Cai Y, Strnad M, Liu C. Arabidopsis ABCG14 protein controls the acropetal translocation of root-synthesized cytokinins. Nat. Com-mun. 2014; 5: 3274.

[19] Kim S, Nie H, Jun B, Kim J, Lee J , Kim S, Kim E, Kim S. Functional genomics by integrated analysis of transcriptome of sweet potato (Ipomoea batatas (L.) Lam.) during root formation. Genes & Genomics.2020; 42: 581–596.

[20] Liu R, Cai XT, Zhao PX, Xu P, Xiang CB. Arabidopsis ERF109 regulates auxin transport-related genes in root development. BioRxiv. 2019.

https://doi.org/10.1101/725572

[21] Ohashi-Ito K, Matsukawa M, Fukuda H. A typical bHLH Transcription Factor Regulates Early Xylem Development Downstream of Auxin. Plant and Cell Physiology. 2013; 54(3): 398–405. https://doi.org/10.1093/pcp/pct013

[22] Guo HS, Xie Q, Fei JF, Chua NH. MicroRNA directs mRNA cleavage of the transcription factor NAC1 to downregulate auxin signals for Arabidopsis lateral root development. The Plant Cell. 2005; 17: 1376–1386.

[23] Nakano Y, Yamaguchi M, Endo H, Rejab NA,Ohtani M. NAC-MYB-based transcriptional regulation of secondary cell wall biosynthesis in land plants. Front. Plant Sci. 2015; 6: 288-31.

[24] Wang H, Avci U, Nakashima J, Hahn MG, Chen F, Dixon RA. Mutation of WRKY transcription factors initiates pith secondary wall formation and increases stem biomass in dicotyledonous plants. Proc. Natl. Acad. Sci. U.S.A. 2010; 107: 22338-22343.

[25] Silverstone AL, Ciampaglio CN, Sun T. The Arabidopsis RGA gene encodes a transcriptional regulator repressing the gibberellin signal transduction pathway. Plant Cell. 1998; 10: 155–169.

[26] Kazan K. Auxin and the integration of environmental signals into plant root development. Annals of Botany. 2013; 112(9): 1655–1665doi:10.1093/aob/mct229

[27] Zhu MD, Xie HJ, Wei XJ, Dossa K, Yu YY, Hui AZ, Tang GH, Zeng XS, Yu YH, Hu PS, Wang JL. WGCNA analysis of salt-responsive core transcriptome identifies novel hub genes in rice. Genes. 2019; 10: 719, doi: 10.3390/genes10090719

[28] Togari Y. A study in the tuberous-root formation of sweet potato. Bull. Natl. Agric.Exp. Sta. 1950; 68:1–96.

[29] Kokubu T. Thremmatological studies on the relationship between the structure oftuberous root and its starch accumulating function in sweet potato varieties. Bull. Fac.Agric. Kagoshima Univ. 1973; 23:1–126.

[30] Siebers T, Catarino B, Agusti J. Identification and expression analyses of newof potential regulators xylem development and cambium activity in cassava (Manihot esculenta). Planta. 2017; 245:539–548.doi: 510.1007/ s00425-00016-02623-00422

[31] Leple´ JC, Dauwe R, Morreel K, Storme V, Lapierre C, Pollet B, Naumann A, Kang KY, Kim H, Ruel K, Lefe`bvre A, Joseleau JP, Grima-Pettenati J, De Rycke R, Andersson-Gunnera S, Erban A, Fehrle I, Petit-Conil M, Kopka J, Polle A, Messens E, Sundberg B, Mansfield SD, Ralph J, Pilate G, Boerjan W. Downregulation of cinnamoyl-coenzyme A reductase in poplar: multiple-level phenotyping reveals effects on cell wall polymer metabolism and structure. Plant Cell. 2007; 19:3669–3691.

[32] Giordano A, Liu Z, Panter SN, Dimech AM, Shang Y, Wijesinghe H, Fulgueras K, Ran Y, Mouradov A, Rochfort S. Reduced lignin content and altered lignin composition in the warm seasonforage grass Paspalum dilatatum by down-regulation of a Cinnamoyl CoA Reductase Gene. Transgenic Res. 2014；23：503–517.

[33] Kim Y, Kim CY, Song W, Park D, Kwon S, Lee H, Bang J, Kwak S. Overexpression of sweetpotatoswpa4 peroxidase results in increased hydrogen peroxide production and enhances stress tolerance intobacco. Planta. 2008; 227: 867–881.

[34] Blee KA, Choi JW, O’Connell AP, Schuch W, Lewis NG, Bolwell GP. A lignin-specific peroxidase in tobacco whose antisense suppression leads to vascular tissue modification. Phytochemistry. 2003; 64: 163–176.

[35] Wang GL, Huang Y, Zhang XY, Xu ZS, Wang F, Xiong AS. Transcriptome-based identification of genes revealed differential expression profiles and lignin accumulation during root development in cultivated and wild carrots。 Plant Cell Rep. 2016; 35:1743–1755. DOI 10.1007/s00299-016-1992-0

[36] Karpinska B, Karlsson M, Srivastava M, Stenberg A, Schrader J, Sterky F, Bhalerao R, Wingsle G. MYB transcription factors are differentially expressed and regulated during secondary vascular tissue development in hybrid aspen.Plant Molecular Biology. 2004; 56: 255–270.

[37] Aloni R, Aloni E, Langhans M, Ullrich CI. Role of cytokinin and auxin in shaping root architecture: regulating vascular differentiation, lateral root initiation, root apical dominance and root gravitropism. Ann Bot. 2006; 97:883–893.

[38] Tamaoki D, Karahara I, Nishiuchi T, Wakasugi T, Yamada K, Kamisaka S. Involvement of auxin dynamics in hypergravity-induced promotion of lignin-related gene expression in Arabidopsis inflorescence stems. Journal of Experimental Botany. 2011; 62(15):5463–5469. https://doi.org/10.1093/jxb/err224.

[39] Herrero J, Carrasco AE, Zapata JM. Arabidopsis thaliana peroxidases involved in lignin biosynthesis: in silico promoter analysis and hormonal regulation. Plant Physiology and Biochemistry. 2014; 80: 192-202.

[40] Steinwand BJ, Xu S, Polko JK, Doctor SM, Westafer M, Kieber JJ. Alterations in auxin homeostasissuppress defects in cell wall function. PLoS ONE. 2014; 9:e98193

[41] Singh V, Sergeeva L, Ligterink W, Aloni R, Zemach H, Doron- Faigenboim A, Yang J, Zhang P, Shabtai S, Firon N. Gibberellin Promotes Sweetpotato Root Vascular Lignification and Reduces Storage-Root Formation. Front. Plant Sci. 2019; 10:1320. doi: 10.3389/fpls.2019.01320

[42] Ma R, Xiao Y, Lv Z, Tan H, Chen R, Li Q, Chen J, Wang Y, Yin J, Zhang L, Chen W. AP2/ERF transcription factor, Ii049, positively regulates lignan biosynthesis in isatis indigotica through activating salicylic acid signaling and lignan/lignin pathway genes. Front. Plant Sci. 2017; 8:1361. doi: 10.3389/fpls.2017.01361

[43] Guo W, Jin L, Miao Y, He X, Hu Q, Guo K, Zhu L, Zhang X. An ethylene response-related factor, GbERF1-like, from Gossypium barbadense improves resistance to Verticillium dahliae via activating lignin synthesis. Plant Mol Biol. 2016.DOI 10.1007/s11103-016-0467-6

[44] Lin JS, Lin CC, Lin HH, Chen YC, Jeng ST. MicroR828 regulates lignin and H2O2 accumulation in sweet potato on wounding. New Phytologist. 2012; 196: 427–440.doi: 10.1111/j.1469-8137.2012.04277.x

[45] Villordon AQ, La Bonte DR, Firon N, Kfir Y, Pressman E, Schwartz. Characterization of adventitious root development in sweetpotato. Hort Science. 2009; 44: 651–655.

[46] Luo Y, Ding N, Shi X, Wu Y, Wang R, Pei L, Xu R, Cheng S, Lian Y, Gao J, Wang A, Cao Q, Tang J. Generation and comparative analysis of full-length transcriptomes in sweetpotato and its putative wild ancestor I. trifida. bioRxiv. 2017, doi: https://doi.org/10.1101/112425

[47] Li B. Dewey CNJBb: RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics. 2011;12(1):323.

[48] Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010; 11:106.

[49] Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008; 9:559. doi:10.1186/1471-2105-9-559

[50] Jeong H, Mason SP, Barabasi AL, Oltvai ZN. Lethality and centrality in protein networks. Nature. 2001; 411: 41–42.

[51] Qin Z, Hou F, Li A, Dong S, Huang C, Wang Q, Zhang L. Comparative analysis of full-length transcriptomes based on hybrid population reveals regulatory mechanisms of anthocyanin biosynthesis in sweet potato (Ipomoea batatas (L.) Lam). BMC Plant Biology. 2020; 20:299. https://doi.org/10.1186/s12870-020-02513-1

[52] Dobrev PI, Vankova R. Quantification of abscisic acid, cytokinin, and auxin content in salt-stressed plant tissues. Plant salt tolerance: methods and protocols. 2012; 251–261.

[53] Djilianov DL, Dobrev PI, Moyankova DP, Vankova R, Georgieva DT, Gajdošová S, Motyka V. Dynamics of endogenous phytohormones during desiccation and recovery of the resurrection plant species Haberlea rhodopensis. Journal of Plant Growth Regulation. 2013; 32: 564. https://doi.org/10.1007/s00344-013-9323-y.

[54] Van-Acker R, Vanholme R, Storme V, Mortimer JC, Dupree P, Boerjan W. Lignin biosynthesis perturbations affect secondary cell wall composition and saccharification yield in Arabidopsis thaliana. Biotechnology for Biofuels. 2013; 6: 46. https://doi.org/10.1186/1754‐6834‐6‐46

Table 1 The data quality of raw sequencing

Sample name	Raw reads	Clean reads	Clean bases	Error rate (%)	Q20(%)	Q30 (%)	GC content(%)
J29_D3_1	50772238	50324636	7.55G	0.03	96.45	90.79	47.90
J29_ D3_2	50727116	50188652	7.53G	0.03	97.07	92.06	48.01
J29_ D3_3	46664608	46042656	6.91G	0.03	97.10	92.09	48.02
J29_ D2_1	65082028	63767890	9.57G	0.03	97.07	92.06	46.55
J29_ D2_2	60019052	59290052	8.89G	0.03	97.04	91.97	46.89
J29_ D2_3	66871460	66124890	9.92G	0.03	96.54	90.95	47.24
J29_ D1_1	55777736	55118700	8.27G	0.03	96.71	91.31	46.83
J29_ D1_2	50937224	50213938	7.53G	0.03	97.17	92.28	47.09
J29_ D1_3	51760190	51079884	7.66G	0.03	96.93	91.75	46.78
J25_ D3_1	47903750	47252134	7.09G	0.03	97.15	92.19	47.16
J25_D3_2	44751504	44212590	6.63G	0.03	97.22	92.30	47.12
J25_ D3_3	55063564	54453240	8.17G	0.03	97.22	92.34	47.21
J25_ D2_1	55564748	54855850	8.23G	0.03	96.73	91.28	46.78
J25_ D2_2	60047924	59300692	8.9G	0.03	97.21	92.37	47.12
J25_ D2_3	49318046	48714970	7.31G	0.03	97.25	92.40	46.87
J25_ D1_1	52140394	51442220	7.72G	0.03	97.16	92.22	46.94
J25_ D1_2	46680802	46129662	6.92G	0.03	97.15	92.21	47.04
J25_ D1_3	54866800	54167484	8.13G	0.03	96.76	91.38	46.85

Table 2 Unigenes annotation information

Database	Unigene number	Percentage (%)
NT	20114	95.11
NR	19137	90.49
KEGG	18849	89.13
SwissProt	16083	76.05
GO	12949	61.23
Pfam	12949	61.23
KOG	11585	54.78
at least one Database	20409	96.50558
all Databases	8885	42.01343

Table 3 DEGs in plant hormone signal transduction pathway

Comparision name	Up	Down	DEGs number
J25-D3vsJ25-D2	24	84	108
J25-D2 vs J25-D1	52	47	99
J29-D3vs J29-D2	8	37	45
J29-D2vsJ29-D1	0	0	0
J29-D1vsJ25-D1	49	64	113
J29-D2vsJ25-D2	12	21	33
J29-D3vsJ25-D3	49	50	99

No competing interests reported.

Comparative Transcriptome Analysis of Storage Root Reveals Regulatory Mechanisms of Lignin Biosynthesis During Root Development in Two Sweetpotato Cultivars

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1